公开(公告)号：US20250148748A1

公开(公告)日：2025-05-08

申请号：US19016900

申请日：2025-01-10

Applicant: Samsung electronics Co., LTD.

Inventor： Praful MATHUR ,  Shashi Kumar Parwani ,  Darshana Venkatesh Murthy ,  Roopa Kotiganahally Sheshadri ,  Aman Sharma ,  Veerendra K Shetty ,  Sunmin Park

IPC: G06V10/26 ,  G06V10/22 ,  G06V10/32 ,  G06V10/44 ,  G06V10/56 ,  G06V10/82

Abstract: Provided are a method and electronic device for interactive image segmentation. The method includes receiving one or more user inputs for segmenting at least one object from among a plurality of objects in an image. The method includes generating a unified guidance map indicates the at least one object to be segmented based on the one or more user inputs. The method includes generating a complex supervision image based on the unified guidance map. The method includes segmenting the at least one object from the image by passing the image, the complex supervision image and the unified guidance map through an adaptive Neural Network (NN) model. The method includes storing the at least one segmented object from the image. Further, the method include configuring the parameters of an adaptive Neural Network based on color complexity analysis, edge complexity analysis and geometry complexity analysis.

公开(公告)号：US12266055B2

公开(公告)日：2025-04-01

申请号：US18154415

申请日：2023-01-13

Applicant: Samsung Electronics Co., Ltd.

Inventor： Basavaraja Shanthappa Vandrotti ,  Ankit Dhiman ,  Akshunn Jindal ,  Shashi Kumar Parwani

IPC: G06T17/20 ,  G06T7/12 ,  G06T7/62 ,  G06T7/70 ,  G06V10/25 ,  G06V10/46

Abstract: A method and system for generating three-dimensional (3D) model augmented related objects from a scene are provided. The method includes creating one or more 3D objects and placing the 3D objects into the 3D scene. Embodiments herein disclose methods and systems for generating 3D augmented reality (AR) objects from a scene. The method may capture an object from the scene, perform a coarse semantic segmentation on the identified object, derive connected contour, generate intermediate contour from at least one outer and inner contours, and configure three-dimensional mesh and texture mapping to generate a three-dimensional model of the captured object.

公开(公告)号：US11948276B2

公开(公告)日：2024-04-02

申请号：US17150800

申请日：2021-01-15

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Shubham Rustagi ,  Sathish Cherukuri ,  Shashi Kumar Parwani ,  Mineni Niswanth Babu ,  Prasenjit Chakraborty

IPC: G06T5/00 ,  G06V20/40 ,  G11B27/031

CPC classification number: G06T5/002 ,  G06V20/49 ,  G11B27/031 ,  G06T2200/24 ,  G06T2207/10016

Abstract: A computer-implemented method includes obtaining, from a storage, a video to be enhanced, based on a selection of a user; determining corresponding sets of object scores for a plurality of objects identified in the video, respectively, based on a set of predetermined factors; identifying a primary object and one or more secondary objects in the video, among the plurality of objects, based on the corresponding sets of object scores; and applying at least one visual effect to the primary object and at least one secondary object, from the one or more secondary objects, in at least a portion of the video, for obtaining an enhanced video with the at least one visual effect applied at least to the portion of the video.